This controlled study proposes to assess the relative effects of home care and institutional care on neuropsychiatrically impaired children and to demonstrate that utilization of appropriate community treatment modalities can significantly reduce state hospital admissions for these children. Children will be randomly assigned to Home vs. Hospital groups as they are deemed eligible for the study for a treatment period of six months. Pretest and post-test measures to assess the effects of either disposition on 1) behavior change in patient, sibs, and peers, 2) parental psychopathology and role function, 3) educational progress, and 4) cost analysis will be made. The following instruments will be used: WISC, Metropolitan Achievement Tests, Conners Behavior Rating Scale, Devereaux Elementary School Behavior Rating Scale, Bender, Psychiatric Status Schedule (for parents), and experimental inventories of medical history and family functioning. Follow up procedures will be instituted to assess the long term impact of institutionalization on these children.